通过去除边缘缓解大脑模拟中的关键节点

IF 4.4 2区 计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE
Yubing Bao , Xin Du , Zhihui Lu , Jirui Yang , Shih-Chia Huang , Jianfeng Feng , Qibao Zheng
{"title":"通过去除边缘缓解大脑模拟中的关键节点","authors":"Yubing Bao ,&nbsp;Xin Du ,&nbsp;Zhihui Lu ,&nbsp;Jirui Yang ,&nbsp;Shih-Chia Huang ,&nbsp;Jianfeng Feng ,&nbsp;Qibao Zheng","doi":"10.1016/j.comnet.2024.110860","DOIUrl":null,"url":null,"abstract":"<div><div>Brain simulation holds promise for advancing our comprehension of brain mechanisms, brain-inspired intelligence, and addressing brain-related disorders. However, during brain simulations on high-performance computing platforms, the sparse and irregular communication patterns within the brain can lead to the emergence of critical nodes in the simulated network, which in turn become bottlenecks for inter-process communication. Therefore, effective moderation of critical nodes is crucial for the smooth conducting of brain simulation. In this paper, we formulate the routing communication problem commonly encountered in brain simulation networks running on supercomputers. To address this issue, we firstly propose the Node-Edge Centrality Addressing Algorithm (NCA) for identifying critical nodes and edges, based on an enhanced closeness centrality metric. Furthermore, drawing on the homology of spikes observed in biological brains, we develop the Edge Removal Transit Algorithm (ERT) to reorganize sparse and unbalanced inter-process communication in brain simulation, thereby diminishing the information centrality of critical nodes. Through extensive simulation experiments, we evaluate the performance of the proposed communication scheme and find that the algorithm accurately identifies critical nodes with a high accuracy. Our simulation experiments on 1600 GPU cards demonstrate that our approach can reduce communication latency by up to 25.4%, significantly shortening simulation time in large-scale brain simulations.</div></div>","PeriodicalId":50637,"journal":{"name":"Computer Networks","volume":null,"pages":null},"PeriodicalIF":4.4000,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Mitigating critical nodes in brain simulations via edge removal\",\"authors\":\"Yubing Bao ,&nbsp;Xin Du ,&nbsp;Zhihui Lu ,&nbsp;Jirui Yang ,&nbsp;Shih-Chia Huang ,&nbsp;Jianfeng Feng ,&nbsp;Qibao Zheng\",\"doi\":\"10.1016/j.comnet.2024.110860\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Brain simulation holds promise for advancing our comprehension of brain mechanisms, brain-inspired intelligence, and addressing brain-related disorders. However, during brain simulations on high-performance computing platforms, the sparse and irregular communication patterns within the brain can lead to the emergence of critical nodes in the simulated network, which in turn become bottlenecks for inter-process communication. Therefore, effective moderation of critical nodes is crucial for the smooth conducting of brain simulation. In this paper, we formulate the routing communication problem commonly encountered in brain simulation networks running on supercomputers. To address this issue, we firstly propose the Node-Edge Centrality Addressing Algorithm (NCA) for identifying critical nodes and edges, based on an enhanced closeness centrality metric. Furthermore, drawing on the homology of spikes observed in biological brains, we develop the Edge Removal Transit Algorithm (ERT) to reorganize sparse and unbalanced inter-process communication in brain simulation, thereby diminishing the information centrality of critical nodes. Through extensive simulation experiments, we evaluate the performance of the proposed communication scheme and find that the algorithm accurately identifies critical nodes with a high accuracy. Our simulation experiments on 1600 GPU cards demonstrate that our approach can reduce communication latency by up to 25.4%, significantly shortening simulation time in large-scale brain simulations.</div></div>\",\"PeriodicalId\":50637,\"journal\":{\"name\":\"Computer Networks\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":4.4000,\"publicationDate\":\"2024-10-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computer Networks\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1389128624006923\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Networks","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1389128624006923","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
引用次数: 0

摘要

大脑模拟有望促进我们对大脑机制的理解、大脑启发智能以及解决大脑相关疾病。然而,在高性能计算平台上进行大脑仿真时,大脑内部稀疏且不规则的通信模式会导致仿真网络中出现临界节点,进而成为进程间通信的瓶颈。因此,有效控制临界节点对于大脑模拟的顺利进行至关重要。在本文中,我们提出了在超级计算机上运行的大脑仿真网络中经常遇到的路由通信问题。针对这一问题,我们首先提出了节点-边缘中心性寻址算法(NCA),基于增强的接近中心性度量来识别关键节点和边缘。此外,我们借鉴在生物大脑中观察到的尖峰同源性,开发了边缘移除转接算法(ERT),以重组大脑仿真中稀疏和不平衡的进程间通信,从而降低关键节点的信息中心性。通过大量仿真实验,我们评估了所提通信方案的性能,发现该算法能准确识别关键节点,且准确率很高。我们在 1600 块 GPU 卡上进行的仿真实验表明,我们的方法可以将通信延迟降低 25.4%,从而大大缩短了大规模大脑仿真的仿真时间。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Mitigating critical nodes in brain simulations via edge removal
Brain simulation holds promise for advancing our comprehension of brain mechanisms, brain-inspired intelligence, and addressing brain-related disorders. However, during brain simulations on high-performance computing platforms, the sparse and irregular communication patterns within the brain can lead to the emergence of critical nodes in the simulated network, which in turn become bottlenecks for inter-process communication. Therefore, effective moderation of critical nodes is crucial for the smooth conducting of brain simulation. In this paper, we formulate the routing communication problem commonly encountered in brain simulation networks running on supercomputers. To address this issue, we firstly propose the Node-Edge Centrality Addressing Algorithm (NCA) for identifying critical nodes and edges, based on an enhanced closeness centrality metric. Furthermore, drawing on the homology of spikes observed in biological brains, we develop the Edge Removal Transit Algorithm (ERT) to reorganize sparse and unbalanced inter-process communication in brain simulation, thereby diminishing the information centrality of critical nodes. Through extensive simulation experiments, we evaluate the performance of the proposed communication scheme and find that the algorithm accurately identifies critical nodes with a high accuracy. Our simulation experiments on 1600 GPU cards demonstrate that our approach can reduce communication latency by up to 25.4%, significantly shortening simulation time in large-scale brain simulations.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Computer Networks
Computer Networks 工程技术-电信学
CiteScore
10.80
自引率
3.60%
发文量
434
审稿时长
8.6 months
期刊介绍: Computer Networks is an international, archival journal providing a publication vehicle for complete coverage of all topics of interest to those involved in the computer communications networking area. The audience includes researchers, managers and operators of networks as well as designers and implementors. The Editorial Board will consider any material for publication that is of interest to those groups.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信